1. Signal Transduction 1.1 Integration of Metabolism and Survival S1.1-1 Regulation of metabolism by estrogen signaling J. Gustafsson, R. Rodrigo and W. Warner Department of BioSciences and Nutrition, Novum, Karolinska Institutet, Stockholm, Sweden. E-mail: jan-ake.gustafsson@mednut.ki.se Estrogen is known to influence glucose homeostasis but the role of estrogen receptors in muscle glucose metabolism is unknown. Therefore, we investigated the expression of the two estrogen receptors, ERa and ERb and their influence on regulation of GLUT4, and its associated structural protein, caveolin-1, in mouse muscle. ERa and ERb are co-expressed in the nuclei of most muscle cells and their levels were not affected by absence of estradiol (in aromatase knockout, ArKO, mice). GLUT4 expres- sion on the muscle cell membrane was not affected by loss of ERb but was extremely reduced in ERa –/– mice and elevated in ArKO mice. Upon treatment of ArKO mice with the ERb agon- ist, DPN, GLUT4 expression was reduced. Caveolin-1 expression was higher in ArKO mice and lower in ERb –/– and ERa –/– mice than in WT littermates. GLUT4 and caveolin-1 were colocalized in WT and ArKO mice, but not in ERb –/– and ERa –/– mice. Thus, ERa is necessary for GLUT4 expression while ERb has a suppressive role. Both ERb and ERa are necessary for optimal caveolin-1 expression. Taken together, these results indicate that reduction in GLUT4 is a plausible explanation for the insulin resistance observed in ERa )/) mice and that colocalization of caveolin-1 and GLUT4 is not an absolute requirement for muscle glucose metabolism. S1.1-2 Nutritional sensing in the MTOR/S6K1 pathway in the development of obesity and diabetes G. Thomas Genome Science, Genome Reserach Institute, Cincinnati, OH, USA. E-mail: thomasg4@uc.edu Earlier we showed that nutrient overload induces S6K1 activa- tion, which suppresses insulin-induced class 1 PI3K signaling leading to insulin resistance. Counter to the prevailing view, our recent studies show that the nutrient, e.g. amino acid (AA), input to S6K1 is not mediated by the tumor suppressor TSC1/TSC2 or its target the oncogene Rheb. In the absence of TSC1/2, we find S6K1 is activated and refractile to insulin, but still regulated by AAs. However this is not the case for Rheb, as siRNA knock- down of Rheb protein levels blocks both the insulin and AA input to S6K1. Nonetheless, withdrawal of AAs has no effect on Rheb-GTP levels, but triggers S6K1 inactivation, indicating Rheb-GTP is necessary but not sufficient for AA induced S6K1 activation. This suggested the AA input to S6K1 is on a parallel pathway to the TSC1/2-Rheb axis. That wortmannin, a class 1 PI3K inhibitor, blocks AA-induced S6K1 activation but AAs do not induce PKB activation, suggested a novel wortmannin sensi- tive signaling component mediated the AA input to S6K1, which we identified as class 3 PI3K, hVps34. In brief, ectopic expression of hVps34 drove S6K1 activation, an effect blocked by hVps34 siRNAs. Moreover, AAs increase hVps34 activity and the pro- duction of PI3P, which serves to recruit FYVE or Px containing proteins to endosomes to build signaling platforms. Consistent with this, S6K1 activation is attenuated by ectopic expression of a dominant interfering cDNA containing two FYVE domains. S1.1-3 Regulation and function of the WNK1 and WNK4 protein kinases D. Alessi MRC Protein Phosphorylation Unit, University of Dundee, Dundee, UK. E-mail: d.r.alessi@dundee.ac.uk The WNK family of protein kinases were originally identified as enzymes that lack a conserved Lys residue normally found in subdomain II of the catalytic domain. Subsequent studies, identi- fied mutations in the genes encoding WNK1 and WNK4, in fam- ilies with an inherited hypertension and hyperkalaemia disorder, called PHAII/Gordon 1 s syndrome. WNK isoforms are large pro- tein kinases (WNK1-2382 residues, WNK4-1243 residues), in which the catalytic domain is located at the N-terminus. Apart from two putative coiled-coil domains, the remainder of the WNK polypeptides possess no obvious structural features. Muta- tions in the WNK1 gene found in PHAII subjects, are deletions in intron-1, which elevate the expression of the WNK1 protein, indicating that hypertension could result from increased expres- sion of WNK1. Consistent with this notion, mice lacking one allele of WNK1, had lower blood pressure. Thus far, the muta- tions in the WNK4 gene found in PHAII subjects, lie distal to both of the putative coiled-coil domains. Little is known about the molecular mechanism by which WNK isoforms regulate cel- lular processes. In my talk I will present our recent results that indicate that the WNK protein kinases are activated by osmotic stress and phosphorylate and activate protein kinases of the STE20 family, termed STE20/SPS1-related Proline-Alanine-rich Kinase (SPAK) and the Oxidative Stress Response kinase-1 (OSR1). Abstracts 4 1.2 Integration of Defence and Survival S1.2-1 Immuno-receptors recognition of antigens: the T-cells case D Gakamsky and I. Pecht Immunology, Weizmann Institute of Science, Rehovot, Israel. E-mail: israel.pecht@weizmann.ac.il Resolution of the interaction mechanism between the T-cell recep- tor (TCR) and its ligands, MHC peptide (MHCp) complexes, is still a central problem in immunology. These are characterized rel- atively high specificity and modest affinities. 3-D structures of TCR-MHCp complexes revealed marked changes are observed, primarily in the TCR, upon ligand binding the. Moreover, both time course and affinity measurements of TCR-MHCp interactions obtained using the Surface Plasmon Resonance (SPR) method established the 0.1–100 mM affinity range for the interactions of the different TCR-MHCp couples examined. These were inter- preted to be a result of relatively slow association (103–105/M/s) and fast complex dissociation (0.1–100/s) rate constants. However, practically all values for the association rate constants were calcu- lated by assuming the operation of a reversible, single step reaction mechanism. This conflicts with the crystallographic evidence for conformational changes taking place in the TCR upon interaction with its ligands, and calls for considering a more complex mechan- ism for MHCp-TCR interaction. Indeed, an in-depth analysis of kinetic data obtained by SPR as well as an independent, FRET based study of TCR-MHCp interactions shows that these fit well two different, though related mechanisms, both involving conform- ational transitions coupled to the binding process. S1.2-2 Amplification and control of B cell antigen receptor signaling M. Reth and Y. Kulathu Institute for Biology III and Max-Planck-Institute for Immunobiology, Freiburg, Germany. E-mail: reth@immunbio.mpg.de Signal transduction from the B cell antigen receptor (BCR) involves not only the activation of BCR-proximal protein tyro- sine kinases (PTK) like Syk and Lyn, but also the inhibition of protein tyrosine phosphatase (PTP) like SHP-1 through the oxi- dation of its catalytic active cysteine. For signaling it is therefore important that the BCR is localized to areas of high production of oxidants and low PTP activity. Under these condition Syk is released from auto-inhibition by binding to the BCR and can amplify the BCR signal through ITAM phosphorylation. The activation of Syk and its release from auto-inhibition involves at the same time a conformational change and its translocation to the plasma membrane via binding of its tandem SH2 domain to phosphorylated ITAMs. To uncouple these two important events in Syk activation, we have targeted Syk permanently to the mem- brane and tested a panel of Syk mutants for their function. These data show that several parts of the Syk molecule are involved in the auto-inhibition control. Once resuming an open conforma- tion, Syk can recruit other signalling molecules thus facilitating their phosphorylation and the propagation of the BCR signal. These studies, which are conducted in a novel signalling pathway reconstitution system with S2 Schneider cells, give a more detailed inside into the mechanistic aspect of BCR signaling. S1.2-3 Temporal regulation of cytolytic T cell programming G. Verdeil, J. Chaix, A. Guimezanes, N. Auphan-Anezin and A. Schmitt-Verhulst Centre d’Immunologie de Marseille-Luminy, CNRS-INSERM- Univ. de la Me ´ diterrane ´ e, Marseille, France. E-mail: verhulst@ciml.univ-mrs.fr Naı ¨ ve CD8 T cells may develop into cytolytic effectors following engagement of their clonally expressed antigen receptor (TCR), thus committing the T cells to a program that includes clonal expansion, differentiation, and long-term survival. Incomplete differentiation has been reported, however, in particular for weak tumor antigens. To understand the molecular basis for the plasti- city in CD8 T cell differentiation, we analyzed the response of monoclonal naı ¨ ve CD8 T cells to antigenic ligands of different affinity. A transient versus stable pattern of gene expression was induced by weak versus strong TCR agonists, respectively. For these transcripts, including those of the IL-2 receptor (IL-2R), cytolytic effector molecules (granzyme B, perforin) and TNF-R family members (GITR, OX40, 4-1BB), expression was sustained upon addition of IL-2, an effect mimicked by expression of an active form of STAT5. Epigenetic modifications of histones asso- ciated with relevant genes were delayed in response to weak agonists. Addition of IL-2 failed to shorten this delay, but induced STAT5 binding to granzyme B and 4-1BB promoters. A delayed and weak ERK1/2 activation in response to weak TCR agonists was correlated with a delayed onset of proliferation. We thus identified the TCR, IL-2R and members of the TNF-R fam- ily, as sequential contributors to the proliferation, differentiation and survival of effector CD8 T cells, involving ERK, STAT5 and NF-kB signaling pathways, respectively. 1.3 Rhythmic Signals: The Setting of Biological Time S1.3-1 An ultradian clock in yeast: metronome for intracellular coherence D. Lloyd 1 and D. Murray 2 1 Microbiology Group (Biosi 1) Cardiff University, Cardiff, Wales, UK, 2 The Systems Biology Instititue, Shinanomachi Research Park, Keio University School of Medicine, Shinjuku-ku, Tokyo, Japan. E-mail: lloydd@cf.ac.uk Self-synchrony of a densely-populated culture of Saccharomyces cerevisiae in aerobic continuous culture under controlled condi- tions of stirring, temperature and pH reveals an ultradian (with a period of about 40\,min) clock. The high amplitude respiratory oscillation is robust in that it can be maintained over extended times(up to months) and its period is compensated for tempera- ture and nutrtional changes. Extensive studies of metabolism and micromolecular synthesis indicate that an oscillatory alternation between oxidative and reductive states underlies the observed outputs (mitochondrial respiratory activity, nicotinamide indica- ated by nucleotides reduced glutathione content, mitochondrial inner membrane potential and levels of transcripts). Ultra- structural changes also occur during redox state transitions; Abstracts 5 mitochondrial matrix volume indicates energization states. We suggest that this timekeeping mechanism serves as the universal coordinating time-base upon which all intracellular reactions, pathways, processes and events are dependent for synchronous coordination and coherence. Circadian control systems that match internal activities to environmental changes may use ultra- dian rhythyms as their bases. The observables of the culture pro- vide us with a unique insight into the temporal organization of the living state in every cell and at all times. Lloyd D. and Murray D. B. Ultradian metronome: timekeeper for orchestration of cellular coherence. TIBS 30: 373–377. S1.3-2 The mammalian circadian timing system: from cyclic transcription to rhythmic physiology B. Kornmann, F. Gachon, H. Reinke, J. Ripperger, G. LeMartelot and U. Schibler Department of Molecular Biology and NCCR Frontiers in Genetics, University of Geneva, Geneva, Switzerland. E-mail: ueli.schibler@molbio.unige.ch Circadian pacemakers were originally believed to exist only in a few specialized cell types, such as SCN neurons. However, in recent years, this view has been challenged by the discovery that self-sustained and cell-autonomous circadian clocks are operative in most peripheral organs. We are studying how such peripheral oscillators can drive overt cycles in physiology. The PAR basic leucine zipper (PAR bZip) proteins DBP, HLF, and TEF serve as clock output regulators. Their expression is governed directly by cellular circadian oscillators. In turn, the rhythmic accumulation of these proteins results in daily cycles of target gene expression. Transcriptome profiling studies with PAR bZip deficient knockout mice revealed that DBP, HLF, and TEF regu- late primarily the transcription of genes involved in hepatic, renal, and intestinal detoxification and genes of the innate and adaptive immune system. REV-ERBalpha, a nuclear orphan receptor acts as a strong transcriptional repressor when bound to RORE elements. It generates circadian expression cycles of BMAL1 and CLOCK, the two positive limb members of the molecular oscillators, by periodically repressing the transcription of these genes. In addition, REV-ERBalpha governs the circadian expression of several enzymes and regulators of lipid metabolism. Our genetic Rev-Erbalpha loss-of-function and gain-of-function studies suggest that circadian transcription in the periphery can be governed by cell-autonomous and systemic cues. S1.3-3 Central and peripheral clocks in ontogeny A. Sumova, Z. Bendova, M. Sladek, R. El-Hennamy, K. Laurinova, Z. Jindrakova and H. Illnerova Department of Neurohumoral Regulations, Institute of Physiology, Academy of Sciences of the Czech Republic. E-mail: sumova@biomed.cas.cz In mammals, endogenous rhythmicity of the principal circadian clock located within the suprachiasmatic nucleus (SCN) is entrained predominantly by a light–dark cycle through pathways emanating from retinal photoreceptors. Via separate pathways, non-photic cues may also reset the clock. The SCN clock entrains circadian clocks in numerous peripheral tissues and controls rhythms of various body functions. During ontogeny, the molecular mechanism responsible for generation of the circadian rhythmicity develops gradually from prenatal to postnatal period both in the SCN and in peripheral organs. At the beginning, the maternal SCN sets the phase of the developing fetal and early postnatal clocks. In altricial rodents, the external light–dark cycle starts to reset the SCN clock only after the first postnatal week. However, full adjustment to the day-length, i.e. to the photope- riod, is accomplished only around the time of weaning. Appar- ently, maternal and photic resetting of the developing circadian clocks employ different and non converging signaling pathways. While the photic entrainment pathways have already been roughly recognized, maternal entrainment pathways are still only suggested. The data will summarize the current knowledge on mechanism of setting the time of the developing SCN and periph- eral clocks. It appears that signaling from mothers and from the outside world to the newborn circadian clocks might be very complex. 1.4 NF-jB Pathway in Normal Physiology and Disease S1.4-1 The IKK complex-linking inflammation to cancer M. Karin Department of Pharmacology, University of California, San Diego La Jolla, CA, USA. E-mail: karinoffice@ucsd.edu A link between inflammation and cancer has been suspected for over two millennia, but its molecular nature remained ill defined. It has also been observed that certain bacterial (for instance Heli- cobacter pylori) and viral (for instance HBV and HCV) patho- gens are major risk factors for certain types of cancer, most notably gastric and liver cancers. We have postulated that tran- scription factor NF-jB may be at the center of this nexus, as NF-jB is activated in response to infection and inflammation and in turn upregulates expression of anti-apoptotic and growth promoting genes. As there are several NF-jB transcription fac- tors, we decided to inactivate the critical catalytic subunit of the IjB kinase (IKK) complex, IKKb, as a way to inhibit activation of most NF-jB forms. While inflammation is a major factor that contributes to the development and progression of CAC and other inflammation-linked cancers and is estimated to be involved in up to 20% of all human cancers, we asked whether inflamma- tion driven by NF-jB has an important role in other forms of cancer where chronic inflammation or infection do not precede tumor development. To that end, we used a model of chemically- induced hepatocellular carcinoma (HCC) based on exposure of mice to a complete and potent carcinogen – diethyl nitrosamine (DEN). Heretofore, DEN administration, although resulting in pronounced cytotoxicity, was not found to trigger an inflamma- tion response. S1.4-2 NF-jB signalling in muscle regeneration N. Rosenthal, F. Mourkioti, E. Lara-Pezzi and M. Pasparakis Mouse Biology Unit, EMBL-Monterotondo, Rome, Italy. E-mail: rosenthal@embl.it The adult mammalian body does retain the robust repair capa- city of the embryo into adulthood and gradually loses its regener- ative potential. Our approach has been to intervene in the mechanisms at work in the mammalian response to damage or Abstracts 6 disease by reducing the impediments to effective regeneration of skeletal muscle. In one intervention, transgenic supplementation of a locally acting Insulin-like Growth Factor 1 isoform (mIGF- 1) promotes efficient tissue repair of damaged skeletal and car- diac muscle without scar formation, and prevents muscle atrophy in heart failure. In a second intervention, repression of the NFjB inflammatory pathway by mIGF-1 in damaged muscle has prompted studies in which mice lacking functional NFjB signal- ling specifically in skeletal muscle exhibit increased muscle regen- erative capacity. In a second intervention, supplemental expression of a naturally occurring, active calcineurin isoform (CnAb1), which is normally induced transiently during skeletal muscle regeneration and in response to mIGF-1 in atrophying muscle, enhances repair and reduced fibrosis in response to dam- age, and activates repressors of the NFjB pathway. Taken together, these observations support the feasibility of recapturing embryonic regenerative capacity by modulating key signalling pathways in the adult to restore injured or degenerating tissues. S1.4-3 The biology of NF-jB interactions with chromatin G. Natoli Department of Experimental Oncology, European Institute of Oncology, Milan, Italy. E-mail: gioacchino.natoli@ifom-ieo-cam- pus.it NF-jB is a family of rapidly activated transcription factors con- served in multi-cellular eukaryotes, whose main role is to control inducible transcription of most inflammatory genes, as well as genes involved in multiple steps of cancer development and pro- gression. Overall, we know far less about the mechanisms regula- ting NF-jB activity in the nucleus than about signaling mechanisms responsible for its activation. Understanding NF-jB-regulated transcription is a crucial task to decode the logic of physiological and abnormal inflammatory responses and it should provide general paradigms about control of inducible transcription in higher eukaryotes. How are NF- jB:DNA transactions regulated? How are NF-jB-dependent genes with different functions expressed with kinetics that suit their function? How different microbes and inflammatory stimuli can induce transcriptional outputs that are plastically adapted to each input? Transcriptional specificity may represent a partic- ularly relevant task in the context of the anti-microbial response, since rapid and successful elimination of any pathogen will require the induction of a suitable transcriptional program. In my talk I will review some recent advancements in this field and their implications. 1.5 Signaling and Cancer: Nuclear Receptor Connection S1.5-1 Androgen action and prostate carcinogenesis F. Saatc¸ iog ˘ lu Division of Cellular and Molecular Biology, Department of Biology, University of Oslo, Oslo, Norway Androgens have critical roles in the development and mainte- nance of the male reproductive system and important for pro- gression of prostate cancer. Using a simple screening protocol, we have cloned androgen responsive genes that are also enriched to prostate for expression. The characterization of three of these genes and the proteins that they encode will be presented: Kallik- rein 4 (KLK4), six transmembrane protein of prostate 1 (STAMP1), and STAMP2. KLK4 belongs to the KLK family, which includes Prostate Specific Antigen (PSA, or KLK3), a secreted protein that is widely used as a diagnostic marker for prostate cancer. KLK4 is regulated by androgens and is highly specific to prostate for expression. Interestingly, KLK4 is the first member of the KLK family that is intracellularly localized. KLK4 is predominantly expressed in the basal cells of the normal prostate gland and overexpressed in prostate cancer. STAMP1 and STAMP2 are six transmembrane proteins. Whereas STAMP1 is largely specific to prostate for expression, STAMP2 has a wider tissue distribution and its expression is exquisitely regulated by androgens. Live cell imaging using STAMPs fused with green flourescent protein (GFP) show that STAMPs shuttle between the trans-Golgi network (TGN) and the plasma mem- brane suggesting that they may be involved in the secretory/endo- cytic pathways. STAMP1 is expressed exclusively in the epithelial cells of the prostate and its expression is significantly increased in prostate tumours compared with normal glands, whereas STAMP2 is highly overexpressed in a subset of the prostate ade- nocarcinoma specimens compared with normal prostate epithelial cells. Ectopic expression studies indicate that these genes may be involved in proliferation of prostate cancer cells. Taken together, these data suggest that KLK4, STAMP1, and STAMP2 contri- bute to the normal biology of the prostate cell, as well as pros- tate cancer progression. In addition to these data, the molecular mechanism of antiandrogen action will be presented, determined by live cell imaging of GFP-tagged androgen receptor (AR) inter- acting with its target sites. S1.5-2 Transcription factor mobility and promoter progression G. L. Hager, R. L. Schiltz, M. Wiench, T. Johnson, S. John, A. Nagaich and Y. Qiu Laboratory of Receptor Biology & Gene Expression, NCI, NIH, Bethesda, MD, USA. E-mail: hagerg@exchange.nih.gov The classical view of nuclear receptor action postulates the static binding of liganded receptors to the promoter. We discovered, however, that nuclear receptors interact dynamically with regula- tory elements in living cells, and have proposed the hit and run hypothesis for receptor function. We have also observed that steroid receptor responsive promoters move through a complex series of activity states, a phenomenon we term promoter pro- gression. Genome-wide profiling of glucocorticoid receptor (GR) regulated loci reveals several classes of response, including genes that are transiently activated and genes that are transiently repressed. Thus receptor action either leads to a series of events Abstracts 7 programmed into each promoter, or the receptor and/or associ- ated factors are subject to a time dependent modification of their activity states. We have discovered that HDAC1 is a coactivator for GR induction of MMTV. A sub-fraction of HDAC1 present in a complex with GR becomes acetylated upon hormone treat- ment. This acetylated form of HDAC1 appears in the GR com- plex when MMTV transcription is inhibited, indicating that acetylation of HDAC1 plays a repressive role on MMTV tran- scription. These findings suggest that a critical component of the GR responsive transcriptional apparatus requires deacetylation by HDAC1 for activity, and inactivation of the enzyme during the induction cycle inhibits this process, thus shutting down the promoter after a period of transient activation. S1.5-3 Analysis of steroid hormone receptor function by gene targeting G. Schu ¨ tz Molecular Biology of the Cell I, German Cancer Research Center, Heidelberg, Germany. E-mail: g.schuetz@dkfz.de Germline and somatic gene targeting of genes for steroid hor- mone receptors allows the characterization of their functions as well as their molecular modes of action. For the glucocorticoid receptor (GR) multiple modes of action have been identified. The receptor activates expression of genes by binding as a dimer to glucocorticoid response elements (GRE) as well as by interaction as a coactivator with DNA-bound Stat5. The receptor is able to repress expression of genes by protein–protein interaction and by binding to negative GREs. Cre/loxP-mediated generation of somatic mutants of the mineralocorticoid receptor (MR) circum- vents the early lethality observed after germline inactivation. Inactivation of MR in the forebrain leads to impaired hippocam- pal-dependent learning, but the limbic MR is dispensable for the maintenance of basal hypothalamic–pituitary–adrenal axis activ- ity. The mechanisms underlying the critical actions of estrogen in the secretion of the gonadotropin-releasing hormone (GnRH) are unknown. A neuron-specific ERa mutation in the forebrain leads to infertility and loss of the positive feedback effects of estrogen upon GnRH neurons. As GnRH neurons do not express ERal- pha, these results indicate that ERalpha-expressing neuronal aff- erents to GnRH neurons are critical for the preovulatory GnRH/ LH surge. These genetic approaches not only reveal novel neural functions of these regulatory molecules in gene expression, but also unprecedented modes of their activity. 1.6 Cell Surface Receptors and Downstream Targets S1.6-1 Small GTPase signalling pathways in tumour biology C. J. Marshall Oncogene Team, Cell and Molecular Biology, Institute of Cancer Research, London, UK. E-mail: chris.marshall@icr.ac.uk Small GTPases of the Ras, Rho and Ral families play important roles in tumour biology. Genetic alterations to small GTPases underscore this role. Ras is mutated in some tumours while wild type RhoA and RhoC are over-expressed particularly in more aggressive tumours. As well as their individual roles it is emer- ging that there are significant interactions between small GTPase signalling pathways. Interesting examples of such interactions are emerging through studies on invasion/cell motility. A number of the transcriptional targets of ERK-MAP kinase signalling down- stream of oncogenic Ras are involved in cell movement and adhesion. Others act through regulating the activation state of Rho-family signalling pathways, for example by down-regulation of ROCK or Rho activation and up-regulating Rac activation. S1.6-2 Signal transduction via receptors for PDGF and TGF-beta – possible targets for tumor therapy C H. Heldin Ludwig Institute for Cancer Research, Uppsala University, Uppsala, Sweden. E-mail: c-h.heldin@licr.uu.se Platelet-derived growth factor (PDGF) and transforming growth factor-beta (TGF-beta) affect cell growth, survival and migration, and have important functions during the embryonal develop- ment. PDGF isoforms exert their cellular effects via two structur- ally similar tyrosine kinase receptors. Since PDGF promotes cell growth and survival, overactivity of the PDGF signaling pathway is associated with diesease, e.g. malignacies. We have explored the use of PDGF antagonists in tumor treatment, and found effi- cient inhibition of tumor growth in animal models of tumors dri- ven by autocrine PDGF production. In addition, we have observed that inhibition of paracrine PDGF stimulation of stro- mal fibroblasts and vessel pericytes lowers tumor interstitial fluid pressure and tumor angiogenesis. TGF-beta has a more compli- cated role in cancer; initially TGF-beta is a tumor suppressor through its ability to inhibit growth and to promote apoptosis of tumor cells. At later stages, when tumor cells become insensitive to the cytostatic effects of TGF-beta, TGF-beta has tumor pro- moter effects through stimulation of epithelial-to-mesenchymal transition of tumor cells, stimulation of angiogenesis and suppression of the immune system. We are currently delineating the signaling pathways involved in the various cellular effects of TGF-beta, and exploring the possible use of TGF-beta antago- nists in tumor treatment. S1.6-3 New functions of focal adhesion kinase: a major Src effector involved in cancer M Frame and B Serrels Beatson Institute, Glasgow, UK. E-mail: m.frame@beatson.gla.ac.uk FAK expression is elevated, often at the level of gene dosage/ amplification, in epithelial cancer cells. We reported the first skin-specific conditional knockout of FAK using Cre-lox technol- ogy. Skin carcinogenesis experiments revealed that FAK defici- ency leads to reduced incidence of papillomas and inhibition of progression to carcinomas. Mechanistically, loss of FAK is asso- ciated with increased apoptosis in both keratinocytes in vitro and in the skin in vivo, particularly in the hair follicle bulge region where the stem cell targets for tumorigenesis reside. We have recently expressed putative gain-of-function and loss-of-function mutants of FAK as skin-targeted transgenes, and have found that elevated FAK expression can cause accelerated carcinogenesis Abstracts 8 and progression to carcinoma. In addition to our in vivo experi- ments, in vitro experiments have identified a new function of the FAK FERM domain in regulating actin assembly via the Arp2/3 complex. Specifically, the FERM domain binds and recruits the Arp2/3 complex to peripheral adhesion sites. Critically, Arp2/3 is subsequently released from complex with FAK by integrin- induced auto-phosphorylation of FAK-tyrosine-397, which lies adjacent to the FAK FERM domain. Moreover, release of Arp2/3 is associated with recruitment of actin into filaments in the vicinity of focal adhesions. These data suggest new ways in which integrin and Src signalling to FAK may contribute to the cancer phenotype. 1.7 Signaling Through Ion Channels S1.7-1 Voltage-gated sodium channel upregulation as an accelerating factor in cancer cell behaviour M.B.A. Djamgoz Divison of Cell & Molecular Biology, Imperial College London, London, UK. E-mail: m.djamgoz@imperial.ac.uk Electrophysiological recordings from contrasting strongly versus weakly metastatic human prostate cancer (PCa) and breast can- cer (BCa) cells showed that the former specifically expressed functional voltage-gated sodium channels (VGSCs). At mRNA level, the upregulation was greater than 1000-fold. Similar upre- gulation of VGSC mRNA and protein was detected in human biopsies. Blocking VGSC activity with the highly specific neuro- toxin, tetrodotoxin (TTX), suppressed the cells’ metastatic behav- iours (MCBs) in vitro. Thus, lateral motility (including galvanotaxis), transverse migration, endocytic membrane activity, adhesion and Matrigel invasiveness were modulated by 50%, consistent with the notion that VGSC activity would potentiate metastasis. Semi-quantitative PCR measurements also revealed that Nav1.7 and Nav1.5 were the culprit VGSCs in PCa and BCa, respectively. Importantly, in both cancers, the VGSC was expressed in its neonatal splice form. In the case of BCa, the neo- natal splice variant differed from the adult in a string of seven amino acids of unique sequence and a polyclonal antibody could be raised to target the VGSC protein. Blocking VGSC activity in the human BCa MDA-MB-231 cell line with either the antibody or siRNA, again, suppressed MCB by 50% and eliminated the TTX sensitivity, i.e. the potentiating effect of VGSC. It is conclu- ded that VGSC expression/activity is a viable novel target for clinical management of metastatic disease. S1.7-2 The role of EAG, a potassium channel, in cancer W. Stu ¨ hmer MBNS, Max-Planck-Institute of Experimental Medicine, Go ¨ ttin- gen, Germany. E-mail: wstuehm@gwdg.de Ion channels are increasingly being linked to cancer and tumour progression. Here we describe a voltage-gated, potassium select- ive channel (EAG or ether-a-go-go) with novel electrophysiologi- cal properties, whose normal physiological function is yet unknown but which acts as an oncogene if expressed ectopicaly. One of the most characteristic properties of the channel, its ionic selectivity, is modulated during cell-cycle transitions. Strikingly, the expression of the human EAG is restricted to brain, but it is also present in several tumour-derived cell lines. While normal tissue is not stained with EAG antibodies, over 75% of tumours tested showed positive staining. Experiments under in vitro condi- tions have demonstrated decreased proliferation of EAG1-expres- sing cells by inhibition of this channel. This inhibition of EAG1 is accomplished using RNA interference, functional anti-EAG1 antibodies, or (unspecific) EAG1 channel blockers. We have used the eXplore Optix system to visualize the distribution of EAG in an in vivo mice tumour model. For this, a specific recombinant Fab fragment against EAG was labelled either with DsRed or Cy5.5. These IR-fluorescent antibody constructs were injected into immuno-suppressed mice carrying previously grafted MDA- MB-435 S human mammary carcinoma cells. This enabled us to follow tumour localization as well as tumour progression. We conclude that EAG is a widely distributed tumour marker with diagnostic and therapeutic potential. S1.7-3 Cyclic nucleotide-gated ion channels A. Menini, A. Boccaccio and S. Pifferi Neurobiology Sector, International School for Advanced Studies, SISSA, Trieste, Italy. E-mail: menini@sissa.it Ion channels activated by the binding of cyclic nucleotides are present in several types of cells. Their main physiological role consists in translating changes in the concentration of cAMP or cGMP into a variation of the membrane potential. Cyclic nuc- leotide-gated (CNG) channels were first discovered in the outer segment of retinal rods and shortly after in the cilia of olfactory sensory neurons. In these sensory systems CNG channels medi- ate sensory transduction by conducting cationic currents carried primarily by sodium and calcium ions. In olfaction the binding of odorants to olfactory receptors in the cilia causes, via G pro- tein activation of adenylyl cyclase, the increase of intracellular cAMP, which directly opens CNG channels. The increase in intracellular calcium entering through CNG channels has both inhibitory and excitatory feedback effects. The opening of Ca-activated Cl channels increases the transduction current because olfactory sensory neurons maintain a high intracellular Cl concentration, whereas calcium in combination with calmodu- lin exerts a negative feedback on CNG channels. By using the whole-cell voltage-clamp technique to record the current, in combination with photolysis of caged cAMP or 8-Br-cAMP, we showed that the Ca feedback on CNG channels is the main molecular mechanism responsible for fast adaptation in olfactory sensory neurons. Abstracts 9 . 1. Signal Transduction 1. 1 Integration of Metabolism and Survival S1 .1- 1 Regulation of metabolism by estrogen signaling J. Gustafsson, R. Rodrigo and. kinase -1 (OSR1). Abstracts 4 1. 2 Integration of Defence and Survival S1.2 -1 Immuno-receptors recognition of antigens: the T-cells case D Gakamsky and I.